Robust Spoof Speech Detection Based on Multi-Scale Feature Aggregation and Dynamic Convolution
Abstract
Spoof speech detection (SSD) can help to protect an automatic speaker recognition system against malicious attacks. However, there exists a great diversity in the spoof utterances generated by different text-to-speech and voice conversion algorithms, resulting in a poor generality of an SSD system to unseen spoofing attacks. To address this problem, we integrate multi-scale feature aggregation (MFA) and dynamic convolution operations into the anti-spoofing framework to detect different local and global artifacts of unseen spoofing attacks. The proposed framework mainly contains eight stacked MFA blocks, where in each block the light-Res2Net module is used to capture multi-scale features and the convolutional kernel is dynamically generated by the local and global statistical information of the inputs. Results on two benchmark datasets (i.e., ADD 2023 Fake Audio Detection and ASVspoof 2021 Logical Access) show the superiority of the proposed method over existing state-of-the-art systems.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Large SSL models have shown very good performance over the deepfake detection task.
“As shown above, large SSL models have shown very good performance over the deepfake detection task.”
From WavLM model ensemble for audio deepfake detection (2024) · cites this paper as [6]
Cites
Tak, H., Todisco, M., Wang, X., Jung, J.-W., Yamagishi, J., & Evans, N. (2022) · The Speaker and Language Recognition Workshop (Odyssey 2022) · doi:10.21437/odyssey.2022-16 · Cited by 237 · see citing works
Kinnunen, T., Lee, K. A., Delgado, H., Evans, N., Todisco, M., Sahidullah, M., Yamagishi, J., & Reynolds, D. A. (2018) · The Speaker and Language Recognition Workshop (Odyssey 2018) · doi:10.21437/odyssey.2018-44 · Cited by 198 · see citing works
The Vicomtech Audio Deepfake Detection System Based on Wav2vec2 for the 2022 ADD Challenge
Martin-Donas, J. M., & Alvarez, A. (2022) · ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp43922.2022.9747768 · Cited by 100 · see citing works
Human Perception of Audio Deepfakes
Müller, N. M., Pizzi, K., & Williams, J. (2022) · Proceedings of the 1st International Workshop on Deepfake Detection for Audio Multimedia · doi:10.1145/3552466.3556531 · Cited by 68 · see citing works
Graph-Based Spectro-Temporal Dependency Modeling for Anti-Spoofing
Chen, F., Deng, S., Zheng, T., He, Y., & Han, J. (2023) · ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp49357.2023.10096741 · Cited by 28 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
WavLM model ensemble for audio deepfake detection
Combei, D., Stan, A., Oneata, D., & Cucu, H. (2024) · The Automatic Speaker Verification Spoofing Countermeasures Workshop (ASVspoof 2024) · doi:10.21437/asvspoof.2024-25 · Cited by 15 · see citing works
“Most approaches use self-supervised models as feature ex-tractors, keeping them frozen during training [2,4,6–8].”
BackgroundMethodInfluential
- Authors
- Wu, H., Zhang, J., Zhang, Z., Zhao, W., Gu, B., & Guo, W.
- Affiliations
- University of Science and Technology of China (CN)
- China Merchants Bank,Shenzhen,China
- Venue
- ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- Publisher
- IEEE
- Published
- 2024-04-14
- Pages
- 10156-10160
- ISSN
- 1520-6149 (ISSN-L)
- OpenAlex
- W4392903636
- Cited by
- 14
- References
- 5
- Citations by year
- 2026: 3 · 2025: 10 · 2024: 1
- FWCI
- 4.13
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE International Conference on Acoustics Speech and Signal Processing (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
- Physical Sciences › Computer Science › Signal Processing › Biometric Identification and Security
- Keywords
- anti-spoofing, multi-scale feature aggregation, dynamic convolution
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
